基礎統計(29)多重線形回帰

📈 クリーンアップ用語

📈 王県住宅情報例

# 예제 데이터
subset = ['AdjSalePrice', 'SqFtTotLiving', 'SqFtLot', 'Bathrooms', 
          'Bedrooms', 'BldgGrade']
print(house[subset].head())          

from sklearn.linear_model import LinearRegression

predictors = [ 'SqFtTotLiving', 'SqFtLot', 'Bathrooms', 
          'Bedrooms', 'BldgGrade']
outcome = 'AdjSalePrice'

house_lm = LinearRegression()
house_lm.fit(house[predictors], house[outcome])

print(f'Intercept: {house_lm.intercept_:.3f}') # 절편
print('Coefficients:') # 계수
for name, coef in zip(predictors, house_lm.coef_):
	print(f'{name}:{coef}')
  
>>>
  
Intercept: -521871.368
Coefficients:
SqFtTotLiving:228.83060360240796
SqFtLot:-0.06046682065307607
Bathrooms:-19442.840398321056
Bedrooms:-47769.95518521438
BldgGrade:106106.96307898083

📈 モデル評価

from sklearn.metrics import r2_score, mean_squared_error
import numpy as np

fitted = house_lm.predict(house[predictors])
RMSE = np.sqrt(mean_squared_error(house[outcome], fitted))
r2 = r2_score(house[outcome], fitted)
print(f'RMSE: {RMSE:.0f}')
print(f'r2:{r2:.4f}')

>>>
RMSE: 261220
r2:0.5406

import statsmodels.api as sm
model = sm.OLS(house[outcome], house[predictors].assign(const=1)) # assign 메서드는 값이 1인 상수 열을 예측변수에 추가한다.(절편을 모델링하기 위해 필요)
results = model.fit()
results.summary()